Have you ever wondered what goes into those computer mod-
els that weather forecasters use to predict the weather? Part sci-
ence, and part art, there is a method to the madness. It’s called Nu-
merical Weather Prediction.
George Santayana, Spanish-American philosopher, once said,
“To know your future you must know your past.” Adapted to weather
prediction, we might say, “You can’t predict the future until you
know what’s going on now.” Computer models might be incredibly
wonderful at what they do, but they can only be as good as the data
that feeds them. Hence, having an accurate knowledge of the cur-
rent state of the atmosphere is just as important as how skillfully the
numerical computer models process the data.
In other words, to predict the weather, you have to know the
current wind, temperature, humidity and pressure in the entire glob-
al atmosphere right now before you can forecast what’s going to hap-
pen in the future. Current weather observations of temperature, pre-
cipitation, and many other mete-
orological elements from the
oceans to the top of the atmos-
phere are gathered.
How do we gather atmos-
pheric data? Satellites are
equipped with sounders that
sense the atmospheric profile
below them. A growing source of
data is commercial and research
aircraft with on-board instru-
ments that sample the atmos-
phere. Twice daily, we take a
snapshot of the earth’s three di-
mensional atmosphere by launch-
ing weather balloons that carry
NUMERICAL WEATHER
PREDICTION 1
WINTER REPORT CARD 4
SPRING OUTLOOK 5
DEPARTING METEOROLOGISTS 6
GOES-17 SATELLITE 7
FORESHADOWING DROUGHT 8
QUARTERLY SUMMARY 9
Inside this issue:
April 2018
N a t i o n a l W e a t h e r S e r v i c e N a t i o n a l W e a t h e r S e r v i c e -- S a n D i e g oS a n D i e g o
Coast to Cactus Weather ExaminerCoast to Cactus Weather Examiner
w e a t h e r . g o v / s a n d i e g o Volume 25 Number 2
Numerical Weather Prediction by Miguel Miller
weather instruments called radiosondes, flying weather stations, that sense the conditions as they
rise above the earth. The collected data is radio-transmitted to receivers on the ground and used
by meteorologists. Radars and wind profilers on land combine with ships and buoys at sea and
commercial and research aircraft in the air to piece together the weather picture. Like a jigsaw
puzzle, all these data pieces are gathered, checked for quality and put together. Once meteorolo-
gists can visualize the current state of the global atmosphere, they’re ready to face the future.
All weather is caused
by atmospheric motions that
can be described by mathe-
matical equations. From
these equations future mo-
tions can be calculated. Nu-
merical weather prediction is
the use of computers to mod-
el the atmosphere and pre-
dict how atmospheric motions
change with time both hori-
zontally and vertically. Many
models use a grid system
where forecast points are laid out in a grid over the area they cover. The distance between cen-
ters of these grids, called grid points, vary with scale and design of the model.
Naturally, the more grid points there are in any model, the finer the resulting detail in the
forecast. However, as the number of grid points increases so does the need for more computing
power. For example, a forecast for 6 x 6 grid (36 forecast points) needs four more times the com-
puting time as a 3 x 3 grid (9 forecast points) even though the actual physical area remains the
same. And that doesn’t even account for the additional forecast time steps. Forecast precision im-
proves at the cost of many more times the number of calculations to produce a forecast for the
same physical area.
The data at midnight
and noon Greenwich Mean
Time (GMT) or Universal time
(UTC or Z), is used by power-
ful computers that model the
atmosphere. The data for
these initial conditions (#1 in
the diagram) is used as input.
The computer uses mathe-
matical equations of motion
to calculate a future state of
the atmosphere in a time step
(#2). The resulting state of
the atmosphere at the end of the time step (#3) then becomes the 'new' initialization input for a
repeat of calculations (#4) for the next time step. This gives the new time step result (#5).
Numerical Weather Prediction—continued
3x3 model grid = 9 forecast points 6x6 model grid = 36 forecast
Graphics on this page from NWS/Jetstream
This process, the result of the pre-
vious time step becoming the initial
input for the next time step, re-
peats itself numerous times until
the end of the model run.
Smaller time step intervals
produce more accurate forecasts as
there is less variation in output at
the end of each computation. But
the cost is that smaller time steps
require more computations. Con-
versely, large time step intervals
require less computation time, but
introduce larger variations in out-
put. Therefore, a tradeoff exists
between time step interval lengths
and grid sizes versus computational
power. As computing power in-
creases in the future, we will be
able to have smaller time step in-
tervals and smaller grid sizes lead-
ing to more detailed and accurate
forecasts. For now, this is why weather models are generally accurate out four or five days. Be-
yond that, differences from one model run to the next begin to show increasing variations in fore-
cast solutions.
The National Centers of Environmental Prediction’s Environmental Modeling Center produc-
es several models with varying scales and grid sizes. There is the Global Forecast Systems (GFS)
suite of models for the entire globe, and several regional models such as the North American
Mesoscale Forecast Systems (NAM) and Rapid Refresh (RAP).
The Environmental Modeling Center
weather models have continually improved in ac-
curacy and will continue to do so in the future.
As computers become faster at processing these
calculations, the grids will become smaller for
better horizontal and vertical resolution. Also
the math used in the calculations will improve as
more data become available to provide better
initialization.
At 122 weather forecast offices across the
country, forecasters receive the computer model
guidance several times a day, applying their hu-
man experience and skill to make the forecast.
Numerical Weather Prediction—continued
The Advanced Weather Interactive Processing System (AWIPS) is the NWS forecaster’s workstation. It displays in real-time current atmos-pheric conditions, including land-based observations, upper air data, and imagery from satellites and radars. It also displays a wide variety of computer models of varying scales depicting numerous meteorologi-cal field variables and derived variables, such as temperature, pres-sure, winds, moisture, instability, vorticity, etc. Forecasters analyze and forecast using computer models generated by numerical weather
prediction. File photo.
Forecasters:
Monitor and analyze the data
Interpret the computer models
Decide what the forecast will be
Prepare the forecast
Transmit forecasts in many for-
mats for specific consumer needs
Precipitation was scarce for California
during this winter overall. Many stations across
the state, especially in the south, recorded one
of the driest winters on record through Febru-
ary. March made a valiant effort to make up for
lost precipitation with some healthy storms. It
was a very tall order and March inevitably fell
short. At least it did lift the seasonal precipita-
tion off the record low floor.
The charts on this page illustrate a pre-
cipitation trace that nearly flat-lined through
December, most of January, and February. They
also show that March storms elevated the precipi-
tation in the Sierra Nevada to the middle two-
thirds category, a decent recovery.
In Southern California, only one signifi-
cant storm in early January kept this winter
from an unprecedented dry end.
Winter Precipitation Report Card
These charts show accumulated precipitation from October 2017 through March 2018. The top charts in each pair compare this season’s La Niña precipitation (black line) to the five strongest La Niña events in his-tory. The bottom charts show this season’s precipita-tion compared to normal and the wet and dry ex-tremes. The charts are for the 8-station Sierra Nevada Index (above right), the Los Angeles Basin (above),
and San Diego County (right).
Before we look ahead at the spring outlook, let’s look back at the winter precipitation out-
look and how well it fared.
The chart at far
left shows the forecast
precipitation for Dec
2017 through Feb 2018
as projected in Novem-
ber 2017. The chart at
near left shows the ob-
served precipitation
anomaly for that time
frame. Overall, the win-
ter precipitation outlook
did quite well, forecast-
ing dry conditions over
most of the south and
wet conditions over much of the north and the Ohio Valley. It did not predict very well the dry
weather that was realized in the Pacific Northwest or the Dakotas.
Spring Outlook The Climate Prediction Center (CPC) re-
leased their Spring Outlook in mid-March. April
through June 2018 are most likely to be warm and
dry across the expanded southwest part of the coun-
try, particularly in Texas and New Mexico.
That warm and dry prognosis will mean wors-
ening drought conditions across the Southwest, and
an expansion of the area where drought is expected
to develop. In contrast, a cooler and wetter season
is forecast to ease drought conditions in the north-
ern plains.
The Spring Outlook by Miguel Miller
What a difference two years makes! On 1 April 2015, there was no snow to measure at Phillips Station in the Sierra Nevada (above). On 30 March 2017, 92 inches of snow and 46.1 inches of SWE were meas-ured (right, above and below). Photos CA
DWR and LA Times.
Departing Meteorologists
It’s spring and promotions are blooming for three meteorologists at our NWS office in San
Diego. While the staff will dearly miss these three, we congratulate them and wish them well as
they move on to the next level of their careers.
Steve Harrison Steve was selected as a lead forecaster at the NWS Forecast office
in San Angelo, Texas. He departed in mid March to his new assignment,
just in time for severe weather season in the southern plains!
Over the past seven years or so, Steve made numerous contribu-
tions to improve service to partners and consumers in Southern California,
particularly in the marine and beach communities. His outreach efforts
greatly improved partner relationships up and down the coast and was
able to fine-tune and improve our services in so many ways. He also
worked on improvements to our warning software that we use in the of-
fice to issue forecasts and hazardous weather products.
We wish Steve the best in his new adventure.
Derek Schroeter Derek is going home. The Maine native was selected to be a me-
teorologist intern at the NWS Forecast office in Gray, Maine.
Derek came to the NWS in San Diego as a Pathways Student in Ju-
ly 2016. He has served admirably in the public service unit, collecting
weather data and answering calls from the public and media. He also
contributed greatly to previous issues of this newsletter in the way of
climate and seasonal prediction patterns. All this he did while complet-
ing his doctorate degree from the University of Delaware. He departed
for his beloved New England in April.
Dr. Schroeter will be a great addition to the forecast team and very likely the climate re-
search community.
Brett Albright Brett was selected as a general forecaster at the NWS Forecast
office in Omaha, Nebraska, and will move there in June. A native of
tornado alley in southeastern Kansas, Brett will be returning to the
land of tornadoes, but just a few hundred miles north.
For the last five years as a meteorologist intern, Brett has been
a catalyst for innovative solutions here at the NWS in San Diego. He
has collected storm reports for the official NWS Storm Data publica-
tion. He has streamlined and improved our methods to prepare and issue weather story graphics,
including the popular “This Day in History” graphics. He has also done extensive education out-
reach. The Omaha office will greatly benefit from Brett’s prodigious talents.
GOES-17 Satellite adapted from a noaa.gov story
NOAA’s GOES-S, the second in a new series of four highly advanced geostationary weather
satellites, blasted into orbit on 1 March 2018 from Cape Canaveral, Florida. GOES-S was renamed
GOES-17 once it established position in a geostationary orbit 22,300 miles above the Earth in mid-
March.
The satellite will pro-
vide faster, more accurate,
and more detailed data in near
real-time to track storm sys-
tems, lightning, wildfires,
coastal fog, and other hazards
that affect the western U.S.,
Hawaii and Alaska.
“The Department of
Commerce and NOAA are mak-
ing good on our commitment to
position the most sophisticated
technology available in space
to provide rapid, accurate, life
-saving weather forecasts,”
said Secretary of Commerce
Wilbur Ross. “Along with its easterly partner GOES-16 (commissioned in 2017), GOES-17 will help
communities and businesses prepare for potentially dangerous weather events and minimize the
hazard to American families and economies.”
Later this year, after undergoing a full checkout and validation of its six high-tech instru-
ments, the new satellite will move to the GOES-West position and become operational. From
there, it will constantly provide advanced imagery and atmospheric measurements, real-time
mapping of lightning activity, and improved monitoring of solar activity and space weather.
In addition to improving weather forecasts, GOES-17 will help forecasters locate and track
wildfires – invaluable information that emergency response teams need to fight fires and evacuate
people who find themselves in harm’s way. GOES-17 will also be an important tool for forecasters
to track and predict the formation and dissipation of fog, which can disrupt airport operations.
“The advanced capabilities of this new satellite will provide vital data to improve forecasts for all
weather hazards across the West and downstream across the remainder of the continental U.S.,”
said Tim Gallaudet, Assistant Secretary of Commerce for Oceans and Atmosphere.
GOES-17 will work in tandem with GOES-16, the first satellite in NOAA’s new geostationary
series, now at the GOES-East position. GOES-17 will extend observational high-resolution satellite
coverage utilizing the revolutionary new technology aboard GOES-16 to most of the Western Hemi-
sphere, from the west coast of Africa to New Zealand, and from near the Arctic Circle to near the
Antarctic Circle. The satellite will provide more and better data than currently available over the
northeastern Pacific Ocean, the birthplace of many weather systems that affect the continental
United States.
All systems are GOES! The successful launch of GOES-S (now GOES-17) takes
place on 1 March in Cape Canaveral, Florida. NOAA photo.
There are a lot of different ways to look at drought, including the very useful U.S. Drought Monitor, prepared by a collaboration of several NOAA agencies (droughtmonitor.unl.edu). While the drought monitor does have some predictive value by simply projecting the future trajectory based on the past precipitation trends, it doesn’t take into account recent and current atmospheric conditions such as temperature, humid-ity and evapotranspiration rates that can help foreshadow drought. That got a group of scientists at NOAA’s Earth System Research Laboratory thinking. Is there a way to see drought conditions coming before they arrive? What role does a thirsty atmosphere play in drought development? So they developed the Evaporative De-mand Drought Index (EDDI), an experimental drought monitoring and early warning guid-ance tool. The tool calculates "the thirst of the atmosphere,” (E0) and compares it to the average condition for a given location and across a time period of interest. It is estimated by the amount of water that would evaporate from the soil and be transpired by plants if the soil were well watered. The EDDI measures the signal of drought using information on the rapidly evolving (daily) conditions of the atmosphere. The EDDI can offer early warning of agricul-tural drought, hydrologic drought, and fire-weather risk by providing near-real-time information on the emergence or persistence of anomalous evaporative demand in a region. A par-ticular strength of EDDI is in capturing the precursor signals of water stress at weekly to monthly time-scales, which makes EDDI a strong tool for prepared-ness for both flash droughts and ongoing droughts. EDDI is a measure of the departure of atmos-pheric thirst (E0) aggregat-ed across a time-window of interest relative to his-torical conditions (from a 30-year climatology). In other words, EDDI takes the current and recent trends in evapo-rative potential (how warm and dry the air is) and compares these trends with the season-al average values. Currently, EDDI is generated daily—though with a 5-day lag-time—by analyzing a da-taset of near-real-time atmospheric conditions. It takes five days to run all the quality control procedures and calculate the evaporative demand. There is also an ongoing effort to forecast EDDI based on seasonal climate-forecast information.
Foreshadowing Drought by Miguel Miller
Check the EDDI trends: https://www.esrl.noaa.gov/psd/eddi
Quarterly Summary by James Brotherton
January High pressure remained in control of the region’s weather during the early portion of January. A powerful winter storm system took aim at Southern California from the 8th to the 10th of January, resulting in areas of heavy rainfall, mountain snow and gusty winds. Most locations received between 1 to 2 inches of rain, with iso-lated amounts of 6-8 inches along favored coastal slopes. Flooding hit Lakeside, Menifee, Hemet, Wrightwood, Cabazon, Highgrove, and Chino. Snowfall was 2 to 8 inches above 6,000 feet with up to 12 inches on the higher peaks. Wind caused major damage to a structure in San Diego. Around 10 swift water rescues were reported in the Inland Empire and San Bernardino County Mountains around Lytle Creek. Three people and a dog were rescued by helicopter from the Santa Ana River Chan-nel in Colton. Several other individuals were swept downstream and rescued in the River-side area. Weather condi-tions remained on the cool side between the 10th to the 12th even though it was dry. The 13th to the 18th was warmer than normal and dry due to upper level high pressure. Light rainfall oc-curred on the 20th and 21st as a Pacific storm system mainly affected locations in Central/Northern California. This storm did bring rough seas (6-9 feet) and gusty winds (20-30 knots and locally to 40) to the coastal waters between the 16th to the 20th. Winds and seas resulted in ten boats being beached near Zuñiga Point (entrance to San Diego Bay) after the boats pulled from their moorings. Two rounds of large long-period northwest swell accompanied the storm and arrived at the beaches between the 16th and 20th of January. This brought high surf to west and northwest facing beaches with peak sets occurring in southern San Diego County. Significant beach erosion was reported along with isolated coastal flooding. Weak low pressure brought cooler weather and very light spotty rain to the region on the 24th and 25th. A surface high built over the Great Basin, resulting in offshore pressure gradients and Santa Ana Winds on the 27th, 28th and 29th. The persistent high pressure and off-shore flow brought above normal temperatures to the region through the end of the month. The east/west nature of the gradients favored San Diego County for the strongest winds. A peak wind gust of 89 mph was reported at Sill Hill.
San Diego - Lindbergh Field Data - January 2018
Max Min Avg Rain
Actual 69.8 53.1 61.5 1.78
Normal 65.1 49.0 57.1 1.98
Anomaly 4.7 4.1 4.4 -0.20
% of normal 90
Max 83 61 1.57
Min 62 47
Lytle Creek got flowing and even flooding after eight inches of rain fell on the
slopes upstream on 9 January. Photo ABC7/Twitter.
Quarterly Summary—continued
Even with the significant storm system early in the month, January precipitation ranged from 50% of normal to 125% of normal. Water year-to-date precipitation ranged from 25% to 50% of normal. This also brought continued above normal temperatures for much of the month across the region. All stations recorded mean temperatures between 4 to 7 degrees above normal, including San Diego Lindbergh Field which was 4.4 degrees above normal.
February High pressure brought dry and mild
weather from the 1st to the 11th.
A weak frontal system on the 12th and
13th brought mainly light scattered showers to
all areas. Although no precipitation occurred
from the 14th to the 18th, conditions remained
on the cool side due to persistent low pressure
trough patterns along the West Coast.
Another couple of light showery rain
events occurred on February 19th into the 20th,
and again on the 23rd and 24th. The final event
of the month was on the 26th and 27th when light rain and showers were the norm, alt-
hough some heavy snow occurred in the mountains. Only one flood event occurred during
the month, on the 27th near Bonita where a road was closed due to flooding.
February precipitation ranged from 50% to 125% of normal. Water year-to-date pre-
cipitation ranged from 25% to 50% of normal.
Temperatures ended up slightly above normal region-wide. All stations recorded
mean temperatures between 0 to 2 degrees above normal, including San Diego Lindbergh
Field which was 1.5 degrees above normal.
San Diego - Lindbergh Field Data - February 2018
Max Min Avg Rain
Actual 67.3 51.5 59.4 0.36
Normal 65.0 50.7 57.9 2.27
Anomaly 2.3 0.8 1.5 -1.90
% of normal 16
Max 81 58 0.32
Min 58 44
Showers off the coast of the Huntington Beach Pier were quite photogenic on 21 February. Photo Bill Cramond.
Quarterly Summary—continued
March Two distinct precipitation events occurred during the month across the region. Some locations received near to above normal precipitation during the month, although some sites continued the trend of below normal precipitation from previous months. A slow moving Pacific Storm System dropped south across California on the 1st through the 3rd, but was not a significant storm. A zonal west to east flow became established on the 4th and continued through the 9th across Southern California. This pattern brought cool but generally dry weather conditions with periods of gusty winds. On the 10th and 11th a weak front brought scattered light precipitation to the area. A wet pattern developed between the 13th and 17th as a series of wet troughs passed through the region. These systems brought heavy precipitation to locations in Central and Northern California but mainly the tail end of these storms which brought widespread moderate precipitation to the
San Diego forecast area. The mountains re-ceived beneficial precipitation of 2 to 6 inches from this series of storms. A colder storm on the 15th brought up to 8 inches of snow to the San Bernardino Mountains. A weak high pressure ridge dominat-ed the weather scene from the 18th to the 20th with milder and dry weather condi-tions during this period. A very intense atmospheric river (also colloquially known as a Pineapple Ex-press) impacted Northern and Central Cali-fornia and eventually brought widespread precipitation to Southern California be-tween the 21st to the 23rd. Amounts were heaviest in the northern areas, especially the San Bernardino Mountains and the
Northern Inland Empire where 1 to 3 inches of rain fell in spots. San Diego County, however, most-ly missed out (a trace at San Diego). A persistent trough (low pressure) pattern occurred from the 24th to the 28th, with cooler than normal conditions and a deep marine layer, but no precipitation, along with several windy periods during this time. A ridge of high pressure developed between the 29th to the 31st which brought a much milder and dry pattern to the region. March precipitation was highly variable across the region. Big Bear Lake managed 183% of normal precipitation, while some sites in the deserts ended up near zero. Water year-to-date pre-cipitation continued to be significantly below normal at the end of March, ranging from just 32% to 54% of normal. Temperatures trended slightly above normal, with mean temperatures between 0 to 2 de-grees above normal, including San Diego which was 1.4 degrees above normal.
San Diego - Lindbergh Field Data - March 2018
Max Min Avg Rain
Actual 67.2 54.3 60.8 0.95
Normal 65.6 53.2 59.4 1.81
Anomaly 1.6 1.1 1.4 -0.86
% of normal 52
Max 74 60 0.36
Min 60 47
Satellite image on 21 March of the decaying atmospheric river
as it stumbled into Southern California. NOAA image.